Considerable effort in recent years has been devoted to the study of shocks in the diffuse interstellar medium. This work has been motivated partly by the observations of rotationally excited states of H2, and partly by the realization that species such as CH(+), OH and H2O might be formed preferentially in hot, post-shock gas. The problem of CH(+) and the difficulties encountered when trying to explain the high column densities, observed along lines of sight to certain hot stars, have been reviewed earlier. The importance of a transverse magnetic field on the structure of an interstellar shock was also demonstrated earlier. Transverse magnetic fields above a critical strength give rise to an acceleration zone or precursor, in which the parameters on the flow vary continuously. Chemical reactions, which change the degree of ionization of the gas, also modify the structure of the shock considerably. Recent work has shown that large column densities of CH(+) can be produced in magnetohydrodynamic shock models. Shock speeds U sub s approx. = 10 km/s and initial magnetic field strengths of a few micro G are sufficient to produce ion-neutral drift velocities which can drive the endothermic C(+)(H2,H)CH(+) reaction. It was also shown that single-fluid hydrodynamic models do not generate sufficiently large column densities of CH(+) unless unacceptably high shock velocities (u sub s approx. 20 km/s) are assumed in the models. Thus, the observed column densities of CH(+) provide a constraint on the mode of shock propagation in diffuse clouds. More precisely, they determine a lower limit to the ion-neutral drift velocity.